Deacidizing gaseous effluents such as, for example, natural gas, synthesis gas, combustion fumes, refinery gas, Claus tail gas, biomass fermentation gas, is generally carried out by washing with an absorbent solution. The absorbent solution allows the acid compounds present in the gaseous effluent to be absorbed. The physico-chemical characteristics of this solution are closely linked with the nature of the gas to be treated: specification expected for the treated gas, selective elimination of a contaminant, thermal and chemical stability of the solution towards the various compounds present in the gaseous effluent to be treated.
Deacidizing of these effluents, notably decarbonation and desulfurization, imposes specific requirements on the absorbent solution:    selectivity towards carbon dioxide in relation to oxygen and nitrogen in the case of fumes, in relation to hydrocarbons in the case of natural gas,    thermal stability,    chemical stability, notably towards the contaminants in the effluent, i.e. essentially oxygen, SOx and NOx, and    low vapor pressure, in order to limit absorbent solution losses at the deacidizing column head.
Currently, the most commonly used solvents are primary, secondary or tertiary aqueous alkanolamine solutions. In fact, the CO2 absorbed reacts with the alkanolamine present in solution according to a reversible exothermic reaction.
An alternative to aqueous alkanolamine solutions is the use of hot carbonate solutions. The principle is based on the absorption of the CO2 in the aqueous solution, followed by the reversible chemical reaction with the carbonates. It is known that addition of additives allows the solvent efficiency to be optimized.
Other decarbonation methods by washing with an absorbent solution such as, for example, refrigerated methanol or polyethylene glycols, are based on a physical absorption of the CO2.
In general terms, the use of all the absorbent solutions described above involves an energy consumption for regeneration of the separation agent that is not insignificant. Regeneration of the absorbent solution is generally carried out by entrainment by a vaporized gas commonly referred to as stripping gas. The thermal energy required for regeneration is split in three parts linked with heating of the absorbent solution between the absorption stage and the regeneration stage (sensible heat of the absorbent solution), its vaporization heat and the binding energy between the absorbed species and the absorbent solution. The binding energy is all the higher as the physico-chemical affinity between the solvent compounds and the acid compounds to be removed is high. In the particular case of alkanolamines, it is more expensive to regenerate a very basic primary alkanolamine such as MonoEthanolAmine than a tertiary amine such as MethylDiEthanolAmine. The vaporization heat of the absorbent solution has to be taken into account since the thermal regeneration stage requires vaporization of a quite significant fraction of the absorbent solution in order to obtain the stripping effect that favours elimination of the acid compounds contained in the absorbent solution. This absorbent solution fraction to be vaporized is proportional to the extent of the association between the absorbed contaminant and the absorbent solution. However, an easily vaporizable absorbent solution is penalized by absorbent solution losses by entrainment upon contact between the gas feed to be treated and the absorbent solution. The part of the sensible heat is essentially linked with the absorption capacity of the absorbent solution: it is in fact proportional to the flow rate of the absorbent solution to be regenerated. The distribution of the energy cost of the regeneration stage between the sensible heat, the vaporization heat and the absorbed gas-absorbent solution binding enthalpy essentially depends on the physico-chemical properties of the absorbent solution and of the absorbed compound.
The goal of the present invention is to provide a gas deacidizing method. The invention proposes decreasing the amount of energy required to regenerate an absorbent solution laden with acid compounds.
The present invention proposes using an absorbent solution that has the property of forming two separable phases when it has absorbed a determined amount of acid compounds: a first phase rich in acid compounds and a second phase poor in acid compounds. This two-phase separation property allows to regenerate only the phase laden with acid compounds.